[Standards-JIG] JEP for storing annotations

Stefan Strigler steve at zeank.in-berlin.de
Thu Sep 30 15:06:11 UTC 2004


Am Do, den 30.09.2004 schrieb Peter Millard um 16:19:

> "To any given JID a user MAY store exactly one annotation to it."
> Using MAY with "exactly one" is kind of strange.

That's true :)

>  I'd recommend
> something like:
> "A user MAY store notes for any contact on their roster.

I don't think there is a need to limit it on contacts in the roster.

>  If a note is
> a stored for a contact, there MUST be only one annotation for that
> contact stored in the 'storage:notes' namespace". Or something
> similar.

How about:
"Annotations are stored in the jabber:iq:private namespace. A storage
element marked by the storage:notes namespace will contain a collection
of 'note' elements, each representing a 'note' to a given entity.
For any given JID there MUST NOT be more than one note."

> It would be nice to show an example of fetching the annotations.

I did this JEP analogously to JEP-0048. I agree there should be an
example how to fetch the annotations. Same for JEP-0048.

> Also,
> I would like to see where (in the login process) annotations should be
> fetched.

Well, I don't know myself. It's left to you (as with JEP-0048).
Unfortunately JEP-0049 doesn't handle concurrency. This is a design flaw
of it.

To make loss of data as unlikely as possible one could agree on fetching
(refreshing) notes each time before displaying a note and storing notes
right after editing a note.

Should I add a remark like that?


> pgm.
> On Thu, 30 Sep 2004 12:56:49 +0200, Stefan Strigler
> <steve at zeank.in-berlin.de> wrote:
> > I've written a JEP now. It's available at [1].
> > 
> > I'm not sure if the XML Schema for this is correct. Perhaps someone with
> > more understanding about how to define XML elements may review it.
> > [1] http://me.in-berlin.de/~zeank/jeps/
> _______________________________________________
> Standards-JIG mailing list
> Standards-JIG at jabber.org
> http://mail.jabber.org/mailman/listinfo/standards-jig

More information about the Standards mailing list